Heat exchangers are critical components in virtually all unit operations involving fluid (gas or liquid) streams. They become even more critical when it is desired to add heat or thermal energy or take away heat or thermal energy from a chemical reaction. For example, endothermic reactions often require, or benefit from, the addition of heat energy. Exothermic reactions, on the other hand, often require, or benefit from, the removal of heat energy. Owing to the economic importance of many such chemical reactions, there is a continual quest for improved performance, both in terms of conversion of reactants to products and in terms of selectivity to desired products relative to undesired products.
MicroChannel Technology (MCT) has been demonstrated to provide many such benefits and recent years have seen a significant increase in the application of MCT to many unit operations. See, e.g., A. A. Rostami et al., Flow and Heat Transfer for Gas Flowing In Microchannels: A Review, 38 Heat and Mass Transfer 359–67 (2002) (applications in medicine biotechnology, avionics, consumer electronics, telecommunications, metrology, and many others) and R. S. Wegeng et al., Compact Fuel Processors for Fuel Cell Powered Automobiles Based on Microchannel Technology, Fuel Cells Bulletin No. 28 (2002) (compact hydrogen generators for fuel cells)., MCT utilizes microchannel devices for carrying out processes that had previously been constrained to far larger equipment; often three to 1,000 times as large for comparable total throughput. MCT devices, which contain features of at least one internal dimension of width or height of less than about 2 mm and preferably less than about 1 mm, have the potential to change unit operations in ways analogous to the changes that miniaturization has brought to computing technology. MCT can be used to advantage in small-scale operations, such as in vehicles or personal (portable) devices. Importantly, too, MCT systems that can be economically mass-produced and connected together to accomplish large-scale operations are very desirable.
More particularly, heat exchangers have become smaller and smaller with more heat energy transferred per unit volume due to the additional area of smaller channels in heat exchangers. Earlier technology includes so-called compact heat exchangers. See, e.g., V. V. Wadekar, a ChE's Guide to CHEs, Chemical Engineering Progress, December 2000, 30–49. Compact heat exchangers provide heat energy transfer rate densities, or heat energy transfer rate per unit volume (thermal power density) (where the volume is the total core volume as defined herein below), only up to about 0.4 W/cc for gas-phase exchangers. MCT heat exchangers, by comparison, provide heat energy transfer rate densities (thermal power density) of about 1 W/cc to 40 W/cc. Compact heat exchangers also have low interstream planar heat transfer percents, typically less than 10 percent. MCT heat exchangers, by comparison, have much higher interstream planar heat transfer percents, typically greater than 10 percent, preferably greater than 20 percent, more preferably greater than 40 percent, and even more preferably greater than 50 percent. In addition, MCT heat exchangers can rely on smaller average approach temperatures when producing the higher thermal power densities.
The above disadvantages of compact heat exchangers can be overcome by the use of MCT heat exchangers. There are problems, however, even with existing MCT heat exchangers. For example, MCT heat exchangers have not been designed which can process more than two separate streams in a single integral device. Processing three or more streams in a heat exchanger can, for example, enable unequal heat gain and loss between the three or more streams. Thus, when it is desirable to transfer heat energy between three or more streams, a compact heat exchanger must be employed or multiple two-stream MCT heat exchangers must be employed. Even multiple two-stream MCT heat exchangers, however, allow significantly more heat transfer to the ambient and the necessary stream transfer piping can cause higher pressure drops to redistribute flows or dead zones and eddies which can cause extended residence times. These extended residence times can cause fouling, corrosion, erosion, decomposition, formation of undesirable byproducts, and, for example, coke can be deposited when processing carbon-containing streams at elevated temperatures. Furthermore, for MCT heat exchangers to realize their full potential, they must be combined in significant numbers to be scaled up to economic large-scale operations. Thus, owing to having a large number of small MCT heat exchangers in close proximity and the close proximity of one channel to another, manifolding the streams entering and exiting an MCT heat exchanger (or any MCT device) becomes a problem.
The manifold design objective is to provide for acceptably uniform flow through a device with an acceptable manifold geometry and stream mechanical energy losses. See, W. M. Kays and A. L. London, Compact Heat Exchangers, 3d ed., at 41 (1984). Restated, manifold design requires tradeoffs among device performance factors as affected by flow uniformity, overall pressure drop, and manifold size and complexity. For example, device performance could be heat transfer performance in the case of endothermic reactions coupled with exothermic reactions within and MCT device. As will be appreciated by those skilled in the art, the manifold design for any given stream is readily approached through application of fluid dynamics. Kays at 41–43.
When manifolding multiple streams in MCT devices, the design problem becomes even greater than designing a two-stream manifold. Having more streams present in a device means proportionally less of the external surface area of that device is available for accessing each stream. The compactness of an MCT device works against the geometric spacing requirements needed to seal manifolds to prevent stream-to-stream leakage. The manifold design must, therefore, address both the design objective stated herein above, as well as the limited external surface area.
Heat exchangers are not the only unit operation to benefit from the push toward miniaturization. Closely related, reactors, too, have begun to shrink in size substantially and with excellent results. Wegneg at 9–12 (vaporizers, reforming reactors, and steam reforming). There remain, however, special problems involving MCT reactors and the need for heat transfer. For example, thermal stresses pose significant problems. MCT devices are manufactured and assembled to much higher tolerances than comparable conventional large-scale devices and multiple MCT devices must be closely-packed to economically match the throughput of comparable to large-scale devices. (An MCT device, while producing high output per core unit volume of the device, typically must be combined in very high numbers to provide comparable throughput.) Thus, temperature differentials that could be easily tolerated by a conventional device of greater dimensions can produce unacceptable thermal stresses in an MCT device which is smaller and thus experiences a much higher temperature gradient. Illustratively, an MCT reactor that is overly constrained geometrically either by multiple integral heat exchangers or integrally-combined multiple integral MCT heat exchanger/reactor units can be subjected to potentially destructive thermal stresses. In general, as a result of the increased efficiency of MCT heat exchangers, they exhibit high temperature gradients with corresponding high thermal stresses. To solve this problem, heat exchangers have been “de-coupled” from the reactors to allow for thermal expansion. In doing so, however, separate piping or tubing is required. As a result, as with multiple two-stream MCT heat exchangers, there can be significant heat loss between multiple units to the ambient and through associated piping or tubing. As noted herein above, such piping connections can become sites for fouling and coke-formation problems. Alternatively, more expensive metals that can tolerate the thermal stresses or inexpensive throwaway devices must be employed.
In addition, the goal of combining multiple heat exchanger/reactor devices to provide economically high total throughput has proved to be elusive. See, e.g., O. Woerz, Microreactors as Tools in Chemical Research, in Microreaction Technology, IMRET 5: Procedings of the Fifth International Conference on Microreaction Technology at 385 (Michael Matlosz et al. eds. October 2001) (“In principle, [it is conceivable that microreactors can also be used for production]. However, serious problems would be encountered.”). In the petroleum processing industry, for example, even minimally-sized specialty units, for, for example, hydrogen production, typically have a capacity of at least one million standard cubic feet per day (scfd) of hydrogen up to about 100 million scfd of hydrogen. A single-stream MCT device, in contrast, produces, at most, 1,000 to 10,000 scfd of hydrogen. Therefore, to provide comparable throughput, a system must comprise from 100 to up to 100,000 closely-integrated arrays of microchannel units.
The present invention overcomes the drawbacks of the prior art of having to provide multiple two-stream heat exchangers with the necessary inter-unit piping, the inability of integrating an MCT heat exchanger with an MCT reactor, and combining a plurality of integrated MCT heat exchanger/reactor devices to form an MCT system to gain the benefits of large-scale operation, that is, high throughput to equal large-scale operations. In doing so, significant thermal power density with multiple streams is achieved, heat loss to the ambient is reduced, corrosion, erosion, decomposition, and coke formation are reduced or eliminated, and higher throughput per unit volume is attained. In addition, thermal stresses are reduced by operating devices with a monotonically increasing temperature profile.